Med OncolDOI 10.1007/s12032-008-9114-7 ORIGINAL PAPERExpression proﬁle of BRCA1 and BRCA2 genes in premenopausalMexican women with breast cancer: clinicaland immunohistochemical correlatesGloria Loredo-Pozos Æ Erwin Chiquete ÆAntonio Oceguera-Villanueva Æ Arturo Panduro ÆFernando Siller-Lopez Æ Martha E. Ramos-Marquez ´ ´Received: 2 August 2008 / Accepted: 15 October 2008Ó Humana Press Inc. 2008Abstract Low BRCA1 gene expression is associated with lower BRCA1 expression than older women (P 0.05).increased invasiveness and inﬂuences the response of BRCA1 and BRCA2 expression correlated in healthy, butbreast carcinoma (BC) to chemotherapeutics. However, not in tumor tissues (TT). Neither BRCA1 nor BRCA2expression of BRCA1 and BRCA2 genes has not been expression was associated with tumor histology, differen-completely characterized in premenopausal BC. We ana- tiation, nodal metastasis or p53 and HER-2 expression.lyzed the clinical and immunohistochemical correlates of After multivariate analysis, only disease stage explainedBRCA1 and BRCA2 expression in young BC women. We BRCA1 mRNA levels in the lowest quartile. Premeno-studied 62 women (mean age 38.8 years) who developed pausal BC has aggressive clinical and molecularBC before the age of 45 years. BRCA1 and BRCA2 mRNA characteristics. Low BRCA1 mRNA expression is associ-expression was assessed by reverse transcriptase-poly- ated mainly with younger ages and advanced clinical stagemerase chain reaction (RT-PCR) and that of HER-2 and of premenopausal BC. BRCA2 expression is not associatedp53 proteins by immunohistochemistry. Body mass index with disease severity in young BC women.(BMI) C27 (52%) and a declared family history of BC(26%) were the main risk factors. Ductal inﬁltrative ade- Keywords BRCA1 Á BRCA2 Á Breast carcinoma Ánocarcinoma was found in 86% of the cases (tumor size Gene expression Á Mexico Á mRNA[5 cm in 48%). Disease stages I–IV occurred in 2, 40, 55,and 3%, respectively (73% implicating lymph nodes).Women aged B35 years (24%) had more family history of Introductioncervical cancer, stage III/IV disease, HER-2 positivity, and Breast carcinoma (BC) currently represents a major healthG. Loredo-Pozos Á F. Siller-Lopez Á M. E. Ramos-Marquez (&) ´ ´ problem worldwide. Depending on the population and ´Instituto de Enfermedades Cronico-Degenerativas. Centro clinical stage, premenopausal women represent 25% of theUniversitario de Ciencias de la Salud, Universidad de people with this type of cancer [1–3]. Many studies haveGuadalajara, Sierra Mojada 950. Colonia Independencia, demonstrated that BC in premenopausal patients has aGuadalajara, Jalisco C.P. 44340, Mexicoe-mail: email@example.com more aggressive clinical course than in older patients [4–6]. In general, breast tumors in young women are bio-E. Chiquete logically different, with a higher proliferation index andDepartamento de Medicina Interna, Hospital Civil de less histological differentiation, when compared with olderGuadalajara ‘‘Fray Antonio Alcalde’’, Guadalajara, Mexicoe-mail: firstname.lastname@example.org patients . Mutations in BRCA1 and BRCA2 genes are responsibleA. Oceguera-Villanueva for 90% of the inherited BC cases . A reduced expression ´Instituto Jaliciense de Cancerologıa, Guadalajara, Mexico of BRCA1 has been associated with increased invasiveness of sporadic or inherited BC ; and as is the case for otherA. Panduro ´Departamento de Biologıa Molecular, Hospital Civil de gene expression markers [10–13], BRCA1 expression inﬂu-Guadalajara ‘‘Fray Antonio Alcalde’’, Guadalajara, Mexico ences the response of BC to chemotherapeutic agents
Med Oncol[14, 15], which might in the future have an impact on slides coated with silane (Sigma Immunchemicals, St.treatment choices . Thus, there is increasing interest in Louis, Missouri, USA). The samples were deparafﬁnised inBRCA1 and BRCA2 expression as alternative biologic xylene and rehydrated via a series of graded alcoholsmarkers of cancer behavior and response to treatment [7, 16– (absolute alcohol to wash away the xylene, followed by18]. However, to the best of our knowledge, the expression 95% and then 70% alcohol, with 5 min duration each forof these genes has been scarcely characterized in premeno- rehydration). Sections were taken to water. Epitopepausal BC patients. In this study, we analyzed the clinical retrieval was carried out in the microwave oven usingsigniﬁcance of expression of BRCA1 and BRCA2 genes in epitope retrieval solution at 95°C for 45 min. Then thepremenopausal women with BC. Our hypothesis was that sections were rinsed under running water and taken to PBStumor mRNA levels of these two genes are associated with buffer for 1 min. Endogenous peroxidase activity wasclinical and laboratory characteristics of disease severity. blocked by incubating the sections in 12 ll of 30% hydrogen peroxide for 30 min, followed by washing in PBS buffer. Then anti-HER-2 antibody at 1:100 dilutionMethods was applied for 30 min followed by rinsing in PBS buffer for 1 min. Visualization reagent was applied for 30 minFrom January 2003 to June 2005, we studied 62 women and rinsed with distilled water, followed by DAB solutionwith BC diagnosed before the age of 45 years, who were for 5 min. DAB was removed with distilled water. Thetreated at the Instituto Jaliciense de Cancerologı´a and at slides were then counterstained with Meyer hematoxylin,the Hospital Civil de Guadalajara ‘‘Fray Antonio Alcalde’’ dehydrated in increasing grades of ethanol, cleared in(Guadalajara Jalisco, Mexico). The internal Committee of xylene, and mounted in Depex. A TT known to react withEthics of our centers approved this study. Informed consent anti-HER-2 antibody was used as positive control. Slideswas obtained either from all patients or their legal proxy. A exposed to PBS without primary antibody were used asstandardized, structured questionnaire was used to collect negative controls in each staining batch. The DAKO Her-data from the patient regarding demography and relevant cep Test Protocol system was used to grade the degree ofantecedents. membrane staining (0 and 1± = negative; 2± = moderate positivity; and 3± = intense positivity).Tissue extraction and histological analysis Immunohistochemical analysis of p53 in tumor tissuesTumor (TT) and healthy tissue (HT) biopsies or mastec-tomies were obtained from each patient (from the same Immunohistochemical analysis of p53 was performedbreast and in the same surgical act, one specimen per according to the manufacturer instructions (Cell Marque,patient) for the histological, gene expression, and immu- Rocklin, California, USA). Brieﬂy, sections of 4 lm thicknohistochemical analyses. These tissues were obtained were cut from appropriately selected parafﬁn blocks con-before exposure to any chemotherapeutic agent and in the taining lesional tissue using the rotary microtome, mountedpremenopausal status. TT and HT were ﬁxed in 10% buf- on prewashed glass slides positively charged and heated atfered neutral formalin, dehydrated in alcohols, cleared in 60°C for 1 h. The samples were deparafﬁnised in xylenexylene, and embedded into low-melting-point parafﬁn, as and rehydrated via a series of graded alcohols twice fordescribed elsewhere [19, 20]. All staining procedures for 4 min each. A 1:10 dilution of epitope retrieval solutionlight microscopy were carried out on 5-lm thick sections was prepared (20x ImmunoDNA Retriever with Citrate,and routine histological examinations were performed for Bio SB) and verted in coupling glasses containing samplesall tissue samples on sections stained with hematoxylin and slices and were placed inside a pressure cooker set at loweosin. TT was composed essentially from tumor cells, and pressure at 95°C for 45 min. Slides were cooled at roomsince HT was extracted from the same affected breast, it temperature for 20 min and rinsed with PBS (pH 7.0) forwas composed from non-macroscopically affected mam- 1 min, followed by washing in peroxidase blocker formary gland and some adipose tissue. 20 min. The preparations were incubated with the primary antibody (DO-7, Cell Marque, 0.5 mM concentration) atImmunohistochemical analysis of HER-2 1:400 dilution for 1 h at room temperature and rinsed with(c-erb-B2/c-neu) in tumor tissues PBS buffer for 1 min. Slides were incubated in the Mouse/ Rabbit ImmunoDetector Biotin LinK visualizing systemSections of 4 lm thick were cut from appropriately and Mouse/Rabbit ImmunoDetector HPR Label for 15 minselected parafﬁn blocks containing lesional tissue using a each and rinsed with distilled water followed by DABrotary microtome (Leica RM 2135, Meyer Instruments, solution for 3 min. A TT known to react with anti-p53Houston Texas, USA). Blocks were mounted on glass antibody was used as positive control. Negative controls
Med Oncolwere also included. The preparations were counterstained 61°C for 1 min, and 72°C for 1 min and a ﬁnal extensionwith Meyer hematoxylin for 1 min, dehydrated with a for 5 min at 72°C. Levels of expression of all transcriptsseries of distilled water, 96% ethanol, absolute ethanol, and were quantiﬁed with a photodocumentation system. Wexylene (15 washes in every step). The mounted glasses used water as ‘‘blank reaction’’ since the RT-PCR assaywere then covered and observed with the microscope. and a sample of mammary adipose tissue as negative control since RNA extraction.RNA extraction and quantiﬁcation in tumorand healthy tissues Statistical analysisIsolation of total RNA from TT and HT was carried out Pearson chi-square and Fisher exact tests were used toaccording to the Chomczynski–Sacchi method . assess proportions in nominal variables for bivariate andBrieﬂy, TT and HT were homogenized using a polytron in homogeneity (when more than two variables) analyses. Tothe presence of Trizol. Chloroform was added to separate compare quantitative variables distributed between twoaqueous phase and total RNA was precipitated with iso- groups, Student t test was performed in distributions ofpropanol at 4°C overnight. Quantity and intactness of RNA parametric variables. Pearson correlation was used inwere routinely tested by determining 260/280 absorbance continuous variables. BRCA1 and BRCA2 mRNA levels areanalysis and ethidium bromide ﬂuorescence of RNA elec- reported as relative units with respect to the mRNAtrophoresed in 1% formaldehyde agarose gels. expression of the housekeeping gene GAPDH. No patient was excluded in any comparison analysis. To ﬁnd inde-Expression analyses of BRCA1 and BRCA2 genes pendent predictors of the expression of BRCA1 in thein tumor and healthy tissues lowest quartile, multivariate analyses were constructed by forward stepwise logistic regression, after selection ofBRCA1 and BRCA2 expression in TT and HT was candidate variables by means of bivariate analyses (selec-accomplished essentially as previously described . HT ted if P 0.1). Adjusted odds ratios (OR) with thewas analyzed as external efﬁciency control of the same respective 95% conﬁdence intervals (CI) are provided. Theaffected breast, when compared with gene expression in ﬁtness of the model was evaluated by the Hosmer-Leme-TT. Three reverse transcriptase reactions per patient were show goodness-of-ﬁt test, which was considered as reliablecarried out in order to obtain the corresponding cDNAs. if P [ 0.2. All P-values are two-sided and consideredReverse transcriptase-polymerase chain reaction (RT-PCR) signiﬁcant when P 0.05 in ﬁnal analyses. SPSS versionwas performed using 2 lg of total RNA transcribed in 13.0 for WindowsTM (SPSS Inc., Chicago, IL) was used in0.05 M Tris–HCl pH 8.3, 40 mM KCl, 7 mM MgCl2 all calculations.buffer containing 0.05 mg/ml of random hexamers, 1 mMdNTPs mix 0.05 U/ml RNase inhibitor and 200 U/mlMoloney murine leukemia virus (M-MLV) reverse trans- Resultscriptase. Samples were incubated for 10 min at 70°C andthen for 60 min at 37.5°C. Reverse transcriptase was fur- We studied 62 premenopausal women with BC. Mean agether inactivated by heating the sample tubes at 95°C for was 38.8 years (range 22–45 years). A high body mass10 min. cDNAs were used immediately for reaction or index (BMI) and a declared family history BC were thewere stored at -20°C until use. Gene ampliﬁcations were main risk factors (Table 1). There were 15 (24%) womenperformed in a PCR buffer of 50 mM Tris–HCl pH 9.0, and aged B35 years, who had signiﬁcantly more history of50 mM NaCl containing a mix of 100 mM dNTPs and 1 U cervical cancer, stage III/IV disease (80% vs. 51%,Taq DNA polymerase. Final oligonucleotide primer con- P = 0.048) and HER-2 protein positivity than older womencentration were 10 lM. Primer sequences were BRCA1 (Table 1). The right mammary gland was affected in 61% ofsense 50 -ACAGCTGTCTGGTGCTTCTGTG-30 , antisense cases. In most patients the histological type of BC was50 -CATTGTCCTCTGTCCAGGCATC-30 ; BRCA2 sense inﬁltrating ductal carcinoma, and in near half of the cohort50 -CTTGCCCCTTTCGTCTATTTG-30 , antisense 50 TACG the tumor size in its greatest measurement was [5 cm.GCCCTGAAGTACAGTCTT-30 , GAPDH sense 50 -CGGA Advanced disease was present in the majority of cases; 73%TGCAACGGATTTGGTCGTAT-30 antisense 50 -AGCCTT with affection to lymph nodes. Well-differentiated tumorsCTCCATGGTGGTGAAGAC-30 . The pair of primers were not observed (Table 1). HER-2 protein expression wasfor BRCA1, BRCA2, and GAPDH ampliﬁed 106, 350, and present in 16 (26%) tumors: 3 had moderate positivity (2±)567 bp fragments, respectively. PCR was run in an auto- and 13 had intense positivity (3±). p53 was expressed in 32mated thermal cycler with initial denaturation at 95°C for (52%) tumors: 9 were weakly positive, 12 moderately10 min followed by 40 cycles each at 95°C for 15 sec, positive, and 11 were highly positive.
Med OncolTable 1 Main characteristics of the study cohortVariables All patients Persons aged B35 years Persons aged [35 years P-Value* (n = 62) (n = 15) (n = 47)Risk factors BMI C 27, n (%) 32 (52) 7 (47) 25 (53) 0.66 Declared family history of BC, n (%) 16 (26) 3 (20) 13 (28) 0.74Tumor size 0.33 Greatest diameter 2 cm, n (%) 3 (5) 1 (7) 2 (4) Greatest diameter 2–5 cm, n (%) 31 (50) 5 (33) 26 (55) Greatest diameter [5 cm, n (%) 28 (45) 9 (60) 19 (40)Tumor histology 0.82 Inﬁltrating ductal, n (%) 54 (87) 13 (87) 41 (87) Lobular, n (%) 3 (5) 1 (7) 2 (4) Medullary, n (%) 2 (3) 0 (0) 2 (4) Mixed, n (%) 3 (5) 1 (7) 2 (4)Histological grade** 0.46 Poorly differentiated, n (%) 12 (19) 4 (27) 8 (17) Moderately differentiated, n (%) 50 (81) 11 (73) 39 (83) Well differentiated, n (%) 0 (0) 0 (0) 0 (0)Clinical staging 0.03 Stage I, n (%) 1 (2) 1 (7) 0 (0) Stage II, n (%) 25 (40) 2 (13) 23 (49) Stage III, n (%) 34 (55) 11 (73) 23 (49) Stage IV, n (%) 2 (3) 1 (7) 1 (2)Immunohistochemistry Positivity to p53, n (%) 32 (52) 8 (53) 24 (51) 0.89 Positivity to HER-2, n (%) 16 (26) 7 (47) 9 (19) 0.03* P-value for differences between patients aged B35 years and older women; Pearson chi-square or Fisher exact test, as corresponded** Grade of cellular differentiationBC Breast carcinoma; CC Cervical carcinoma; BMI Body mass index Mean BRCA1 (0.583 vs. 0.344 relative units, respec- signiﬁcant differences in mean tumor mRNA levels oftively; P 0.001) and BRCA2 (0.286 vs. 0.149 relative BRCA1 and BRCA2 genes according to other clinical andunits, respectively; P = 0.006) mRNA levels were higher histological characteristics (Figs. 2 and 3). After multi-in TT than in HT. No correlation was observed between TT variate analysis adjusted for differentiation grade,and HT with respect to BRCA1 or BRCA2 expression. histological type, lymph nodes implication, age, relevantHowever, mRNA levels of BRCA1 correlated with those of antecedents, p53 and HER-2 proteins expression; the onlyBRCA2 in HT, but not in TT (Fig. 1); which denotes a factor independently associated with BRCA1 mRNA levelsdifferential deregulation in mRNA expression of these in the lowest quartile was clinical staging, when consideredgenes in tumor cells. BRCA1 mRNA was not detected in 4 as a continuous variable (OR: 3.22, 95% CI: 1.01–10.27).(6.5%) tumors, and 15 (24%) fell in the lowest quartile ofthe sample. On the other hand, BRCA2 mRNA was nega-tive in 19 (31%) specimens, which coincided exactly with Discussionthe lowest quartile of the sample. TT of women agedB35 years had signiﬁcantly lower levels of BRCA1 gene We found that in premenopausal women with BC a lowexpression, as compared with older women (Fig. 2). Also, BRCA1 gene expression is associated with a younger agewomen aged B35 years had more cases of BRCA1 and clinical staging. This molecular marker was not asso-expression in the lowest quartile of the sample, when ciated with other characteristics of disease severity as p53compared with their older counterparts (53% vs. 19%, or HER-2 expression, grade of differentiation, lymph nodesrespectively; P = 0.002). However, there were no other implication and other factors previously linked to this
Med Oncol developing BC [1, 18]. Race differences and disease severity may also account for the heterogeneity of the results of studies designed to explore the clinical signiﬁ- cance of gene expression proﬁle in deﬁning prognosis in young BC women. A higher than expected frequency of p53 positivity status in TT was found in our cohort (52%; 53% in those aged B35 years and 51% in women aged C 35 years), and a similar picture in HER-2 only in women aged B35 years. Maru, et al.  have reported 50% cases of p53 expres- sion in women aged B30 years; and Gammon, et al. 44% in a cohort of women aged B45 years . In Mexican women it has been reported that a very high frequency of p53 positivity status (as high as 78.5% in young women with advanced disease), a high frequency of lymph nodes involvement and advanced clinical stages, either in pre- menopausal or older women [29, 30]. Thus, it appears that these characteristics are unusual with respect to other populations, but common in a Mexican cohort. Inherited inﬂuence for the development of BC might be a signiﬁcant pathogenic factor in the patients included in our report. With the objectives and methodology applied in the present study, we could not determine the exact role of BRCA1 and BRCA2 gene mutations on disease behavior. Nevertheless, we could expect a high frequency of predisposing gene variants in our patients, as suggested by the family history of BC (26%) and the age cut-off that was considered. mRNA levels of BRCA1 and BRCA2 genes were higher in TT than in HT. This ﬁnding may be attributable to the fact that HT was composed from cells with a relatively low proliferation index. BRCA1 expression correlated with that of BRCA2 in HT, but not in tumors. This ﬁnding may denote a pathological deregulation in mRNA expression of these genes in tumor, but not in healthy cells. Women aged B35 years had signiﬁcantly lower levels of BRCA1 mRNA and they had more cases of BRCA1 expression in theFig. 1 Correlation between BRCA1 and BRCA2 expression in lowest quartile of the sample, as compared with olderhealthy (a) and tumor tissue (b). Relative BRCA1 or BRCA2/GAPDH women. This is in accordance with previous observationsmRNA levels are depicted in y axes. Centered line represents regarding a more ‘‘deleterious’’ gene expression proﬁle incorrelation ﬁt; outliers represent the mean 95% conﬁdence interval young versus older women [9, 16–18, 22]. However, in a multivariate analysis the only factor independently asso-marker [9, 16–18, 22]. Nevertheless, it is not unusual that ciated with BRCA1 mRNA levels in the lowest quartile wasseveral markers are associated with some clinical and clinical staging, even when adjusting for age, either as amolecular variables in several studies, but not, or only to continuous or a discrete (patients aged B35 years versussome extend in other reports, especially when considering older women) variable. These observations satisﬁed ouryoung women [23–27]. These discrepancies could be due hypothesis in that low BRCA1 mRNA levels are indepen-to different sample sizes and age cut-offs used when dently associated with disease severity. On the other hand,selecting patients for analyses. As was shown in the present we could not ﬁnd any relationship between BRCA2 genereport, an age B35 years seems to distinguish among the expression proﬁle and clinical, histological, or molecularyoung premenopausal women, those with the worse clini- variables. Indeed, an important limitation of our study iscal and molecular proﬁle. It is well known that women the small sample size, which may be underpowered toyounger than 35 years had a high frequency on inherited detect some small, but pathophysiologically importantgene variants that render them with a high probability of differences in disease behavior.
Med OncolFig. 2 BRCA1 expression according to several clinical and immu- Filled bars represent arithmetic means; error bars represent standardnohistochemical variables (n = 62, in every comparison). Relative error of means (±2)BRCA1/GAPDH mRNA levels in tumor tissues are depicted in y axes.Fig. 3 BRCA2 expression according to several clinical and immu- Filled bars represent arithmetic means; error bars represent standardnohistochemical variables (n = 62, in every comparison). Relative error of means (±2)BRCA2/GAPDH mRNA levels in tumor tissues are depicted in y axes. In conclusion, young premenopausal women with BC severity in very young women with BC. BRCA2 expressionhave aggressive clinical, histological, and molecular char- seems not to be a good molecular marker of severity inacteristics. BRCA1 expression is a useful marker of disease premenopausal BC.
Med OncolAcknowledgments All the authors have contributed to the concep- 14. Egawa C, Motomura K, Miyoshi Y, Takamura Y, Taguchi T,tion and design of the work and the analysis of the data in a manner Tamaki Y, et al. Increased expression of BRCA1 mRNA predictssubstantial enough to take public responsibility for it; each one favorable response to anthracycline-containing chemotherapy inbelieves the manuscript represents valid work; and each has reviewed breast cancers. Breast Cancer Res Treat. 2003;78:45–50. doi:the ﬁnal version of the manuscript and has approved it for publication. 10.1023/A:1022101310500.The authors had full access to all the data in the study and take 15. Zhou C, Smith JL, Liu J. Role of BRCA1 in cellular resistance toresponsibility for the integrity of the data and the accuracy of the data paclitaxel and ionizing radiation in an ovarian cancer cell lineanalysis. The authors of this article declare that the article is original carrying a defective BRCA1. Oncogene. 2003;22:2396–404. doi:and has not been published or submitted for publication elsewhere, and 10.1038/sj.onc.1206319.that there is no any afﬁliation with any organization with a direct or 16. Al-Mulla F, Abdulrahman M, Varadharaj G, Akhter N, Anim JT.indirect ﬁnancial interest in the subject matter discussed in the man- BRCA1 gene expression in breast cancer. A correlative studyuscript that may affect the reporting of the work submitted. between real-time PCR and immunohistochemistry. J Histochem Cytochem. 2005;53:621–9. doi:10.1369/jhc.4A6544.2005. 17. Rakha EA, El-Sheikh SE, Kandil MA, El-Sayed ME, Green AR, Ellis IO. Expression of BRCA1 protein in breast cancer and itsReferences prognostic signiﬁcance. Hum Pathol. 2008;39:857–65. doi: 10.1016/j.humpath.2007.10.011. 1. Ganz PA, Greendale GA, Petersen L, Kahn B, Bower JE. Breast 18. Klauber-DeMore N. Tumor biology of breast cancer in young cancer in younger women: reproductive and late health effects of women. Breast Dis. 2005–2006;23:9–15. treatment. J Clin Oncol. 2003;21:4184–93. doi:10.1200/JCO. 19. Ramos-Marquez ME, Grijalva G, Armendariz-Borunda J. Duc- 2003.04.196. tular hyperplasia is characterized by an over expression of c-Myc 2. Choi DH, Lee MH, Bale AE, Carter D, Haffty BG. Incidence of in bile duct ligation ? furan injured rats: possible role of inter- BRCA1 and BRCA2 mutations in young Korean breast cancer leukin-6. Hepatol Res. 2002;22:127–38. doi:10.1016/S1386- patients. J Clin Oncol. 2004;22:1638–45. doi:10.1200/JCO.2004. 6346(01)00121-8. 04.179. 20. Salgado S, Garcia J, Vera J, Siller F, Bueno M, Miranda A, et al. 3. Yankaskas BC. Epidemiology of breast cancer in young women. Liver cirrhosis is reverted by urokinase-type plasminogen acti- Breast Dis. 2005–2006;23:3–8. vator gene therapy. Mol Ther. 2000;2:545–51. doi:10.1006/mthe. 4. Walker RA, Lees E, Webb MB, Dearing SJ. Breast carcinomas 2000.0210. occurring in young women (35 years) are different. Br J Cancer. 21. Chomczynski P, Sacchi N. Single-step method of RNA isolation 1996;74:1796–800. by acid guanidinium thyocyanate-phenol-chloroform extraction. 5. Adami HO, Malker B, Holmberg L, Persson I, Stone B. The Anal Biochem. 1987;162:156–9. doi:10.1016/0003-2697(87) relation between survival and age at diagnosis in breast cancer. N 90021-2. Engl J Med. 1986;315:559–63. 22. Baldassarre G, Battista S, Belletti B, Thakur S, Pentimalli F, 6. Chung M, Chang HR, Bland KI, Wanebo HJ. Younger women Trapasso F, et al. Negative regulation of BRCA1 gene expression with breast carcinoma have a poorer prognosis than older women. by HMGA1 proteins accounts for the reduced BRCA1 protein Cancer. 1996;77:97–103. doi:10.1002/(SICI)1097-0142(1996 levels in sporadic breast carcinoma. Mol Cell Biol. 2003;23: 0101)77:197::AID-CNCR16[3.0.CO;2-3. 2225–38. doi:10.1128/MCB.23.7.2225-2238.2003. 7. Armes JE, Trute L, White D, Southey MC, Hammet F, Tesoriero 23. Choi DH, Kim S, Rimm DL, Carter D, Haffty BG. Immunohis- A, et al. Distinct molecular pathogenesis of early-onset breast tochemical biomarkers in patients with early-onset breast cancers in BRCA1 and BRCA2 mutation carriers. A population- carcinoma by tissue microarray. Cancer J. 2005;11:404–11. doi: based study. Cancer Res. 1999;59:2011–7. 10.1097/00130404-200509000-00008. 8. Pherson K, Steel C, Dixon J. Breast cancer-epidemiology, risk 24. Maru D, Middleton LP, Wang S, Valero V, Sahin A. HER-2/neu factors and genetics. BMJ. 2000;321:624–8. doi:10.1136/bmj. and p53 overexpression as biomarkers of breast carcinoma in 321.7261.624. women age 30 years and younger. Cancer. 2005;103:900–5. doi: 9. Thompson ME, Jensen RA, Obermiller PS, Page DL, Holt JT. 10.1002/cncr.20850. Decreased expression of BRCA1 accelerates growth and is often 25. Jmor S, Al-Sayer H, Heys SD, Payne S, Miller I, Ah-See A, et al. present during sporadic breast cancer progression. Nat Genet. Breast cancer in women aged 35 and under: prognosis and sur- 1995;9:444–50. doi:10.1038/ng0495-444. vival. J R Coll Surg Edinb. 2002;47:693–9. ˜10. Munoz-Gonzalez D, Zeichner-Gancz I, Candelaria M, Ramirez- ´ 26. Guerra I, Algorta J, Dıaz de Otazu R, Pelayo A, Farina J. ˜ Ugalde MT, Perez-Sanchez M, Cervantes-Vazquez G, et al. Her- Immunohistochemical prognostic index for breast cancer in 2/neu expression as a predictive factor for response to anthra- young women. Mol Pathol. 2003;56:323–7. doi:10.1136/mp. cycline-based chemotherapy in a Mexican population of locally 56.6.323. advanced breast cancer patients. Med Oncol. 2005;22:23–8. doi: 27. Ibrahim EM, Ezzat AA, Rahal MM, Raja MM, Ajarim DS. 10.1385/MO:22:1:023. Adjuvant chemotherapy in 780 patients with early breast cancer: ´ ´11. Fernandez-Sanchez M, Gamboa-Dominguez A, Uribe N, Garcıa- ´ 10-year data from Saudi Arabia. Med Oncol. 2005;22:343–52. Ulloa AC, Flores-Estrada D, Candelaria M, et al. Clinical and doi:10.1385/MO:22:4:343. pathological predictors of the response to neoadjuvant anthracy- 28. Gammon MD, Hibshoosh H, Terry MB, Bose S, Schoenberg JB, cline chemotherapy in locally advanced breast cancer. Med Brinton LA, et al. Cigarette smoking and other risk factors in Oncol. 2006;23:171–83. doi:10.1385/MO:23:2:171. relation to p53 expression in breast cancer among young women.12. Acharya CR, Hsu DS, Anders CK, Anguiano A, Salter KH, Cancer Epidemiol Biomarkers Prev. 1999;8:255–63. Walters KS, et al. Gene expression signatures, clinicopathologi- 29. Gerson R, Alban F, Villalobos A, Serrano A. Recurrence and cal features, and individualized therapy in breast cancer. JAMA. survival rates among early breast cancer cases with triple nega- 2008;299:1574–87. doi:10.1001/jama.299.13.1574. tive immunophenotype. Gac Med Mex. 2008;144:27–34.13. Kimura M, Koida T, Yanagita Y. A study on telomerase activity 30. Crabtree B, Neme Y, Rivera S, Olivares G. Hormone Receptors, and prognosis in breast cancer. Med Oncol. 2003;20:117–26. doi: HER2/neu and p53 in 1027 Mexican patients with breast cancer. 10.1385/MO:20:2:117. J Clin Oncol. 2005;23:S903.